Testing Schrödinger’s Cat

I’d like to share with you the famous thought experiment “Schrödinger’s Cat”.I’ve known about this for a while, but it popped back in to my head recently and can’t help but find it fascinating!

If you are not already familiar with the Schrödinger’s Cat thought experiment, let me give you a brief summary.

Schrödinger’s Cat: A cat, along with a flask containing a poison and a radioactive source, is placed in a sealed box shielded against environmentally induced quantum decoherence. If an internal Geiger counter detects radiation, the flask is shattered, releasing the poison that kills the cat. The Copenhagen interpretation of quantum mechanics implies that after a while, the cat is simultaneously alive and dead. Yet, when we look in the box, we see the cat either alive or dead, not both alive and dead.

I find this incredibly fascinating. How can something be both alive and dead at the same time? That’s impossible, right? The more I think about this, the more my mind starts to wander, and I couldn’t help myself linking this to Software Testing.

Imagine this scenario. A developer has fixed several defects in a piece of Software and has handed it to you for Testing. Before you start testing the application, are those defects fixed? We can assume that the defects are both fixed and not fixed and that the only way to prove one way or another is to Test it, or to peer into the box.

Testing, to me, is looking inside the box to get rid of any doubt and to prove the outcome one way or another. But isn’t Testing so much more than just looking inside the box? Is the box just the beginning? Do you even need to see inside the box to determine whether the Cat is dead or alive?

Like I say, this really fascinates me and I would have loved to have been hired as a Tester on this project.

Here’s how I would go about it.

The Mission:

To determine whether the Cat inside the box is either dead or alive.


Firstly, I feel the following questions need answering.

  • Define “Dead” and “Alive”
  • What if the Poison was contaminated with the radiation?
  • What if you opened the box just as the hammer hits the poison?
  • What if the Cat knocked the poison over accidentally? Was it securely fastened?
  • What if the Cat is allergic to the box / confined spaces?
  • Was this a fast acting poison? I.e. after the hammer hit the flash, is there a period where the Cat is dying?
  • Where any radioactivity tests performed beforehand to see if the Cat was slightly radioactive before entering the box?
  • Was the Cat suffering from any diseases before being put in the box?
  • Where any structural tests performed on the box itself to prove that external radiation could not enter it?
  • What material is the box made out of? Was the box lead lined? What if the Cat licked its paws and suffered lead poisoning?
  • Was this test carried out multiple times? If not, could we run it multiple times.
  • Was the cat given anything to eat or drink before hand? Was the food tested for radioactivity?
  • Was the equipment calibrated? I.e. would the hammer definitely break the glass?

I could go on, but I would really like to take a look at the equipment used and the condition of the Cat beforehand before performing any tests. There are a lot of factors that could interfere with the results and addressing some of these questions may remove some of the bias in the equipment. For example, if the Cat was fed a certain type of Cat food beforehand that contains slight amounts of radiation, then this could affect the results.

The Cat is in the box, time for Testing:

Schrödinger says the only way to know the fate of the Cat for sure is to lift up the lid and take a peek inside. As Testers, we know that there is more than one way to “look inside” the box.

For example, we could install sensors, meters or microphones and display these results on the outside of the box so that we can clearly see what’s happening inside. Ask for testability and install a window in the side of the box or sound an alarm when the poison has been released. If the alarm is not ringing, the Cat is presumed alive. Have any prior tests been run? Do we have anything to compare against?

I can imagine that this thought experiment would have turned out very differently if Schrödinger had have hired a Software Tester when coming up with this experiment. Can you think of anything else that you could try or anything that you would do differently if faced with this situation?

I’d like to point out that Schrödinger’s never actually carried out this experiment and therefore no animals were harmed in the making of this Blog post.

12 thoughts on “Testing Schrödinger’s Cat

  1. With my luck, when the inside of the box is examined is nothing at all in the box. (or a live tiger) :)

    Q: How do you know it is the same cat?

  2. 1. Put a dog or mouse in the box.
    2. Do I want the cat to be alive?
    3. Shake the box.
    4. Kill the cat yourself.
    5. Replace the cat with another cat regardless of it’s condition
    6. Replace the cat with a dog cause dogs are better.
    7. Change the material of the box to something transparent.
    8. Poke a hole in the box.
    9. Call the cats name.
    10. Bark at the box.
    11. Outsource it to the RSPCA.

    • Bark at the box! lol. Great stuff!

      This is why he should have hired a Software Tester! Apart from finding out for sure if the Cat is dead or Alive, he would have had a good laugh finding out!

  3. Took me some time to get back to posting here.. interesting mind experiment.
    Before more time passed, I just jot down my thought fragments and ideas.

    What is the *most* important part of the mission (for me): Dead or Alive!?
    To the questions: All are be valid to be asked, but only some (imho) are relevant to the mission.
    Define Dead and Alive is the major one for me.

    Also the image shown on the blog post implies two areas in the box. one for the cat, the other for hammer and poison. So the cat could not knock the poison over (at least without rattling the whole contraption a lot).
    But that is me *assuming* the image is relevant to the blog text. :-)

    For some of the questions I would raise a counter-question: Does it matter to the mission, if .. (..the cat is allergic to the box; .. she was poisoned beforehand, etc.)?
    If yes, refine the mission statement or provide more information why it is relevant.
    If not, dont follow up on these questions right now.

    Was this test carried out multiple time? With the same cat? I hope the outcome won’t change, otherwise I should start going to church. ;-)
    Maybe thats where the saying “Don’t flog a dead horse” originated?^^

    One aspect I find missing is the factor of “Time”. When should the state “Dead” be reached?
    It could be infered by the poison used and its regular effect time on an animal the size of the cat (hm, use a bigger cat would then still be a “cat”, but it will take longer to reach its effect). So after X minutes the “measurement” (peeping into the box) should take place and reach a verdict.
    Btw: We didnt talk *how* “dead” is measured. I just realise, I assumed my eyes will give me the result. What about blind observers? How many different ways are there to establish “She is dead, Jim!”? What way should we use? Can they contradict each other (heuristic style)?
    Funny enough there was a twitter discussion a few weeks back about what constitutes “dead” and “not dead” .

    My thoughts to : “look inside” the box => install alarm (test code, test tools) => can influence the outcome => no alarm doesnt necessary mean the cat is still alive or the poison is not released, *IF* the tool fails. Also a window could let the poison leak out the window edges and therefore the effect on the cat is not deadly anylonger (assuming poison quantity is calculated as “just enough” to kill a cat in a small confined space.

    I probably need to take another day off to bring my fragments here in a proper order; still hope some might be useful.


  4. Not to rain on your parade but by the mere fact of testing this experiment you are changing the results of the experiment according to quantum mechanics. the whole point of me going into the box is not to determine if i am dead and alive. the point is to show that if Schrodinger equation is used then according to quantum physics both solutions have been met.

    The act of putting sensors inside the box is the essentially the same as opening the box.

  5. Many of you are suggesting ways to determine wether the cat is alive by adding sensors and technologies, that is not the point of this expermient. It was to explain how any property of any electron cannot be know for sure at any time.

  6. Wait a minute… If you have ‘more than one way to “look inside” the box’ you are still looking inside of the box… Your point seems to be pointless.

    But I do get your point of saying “Testing, to me, is looking inside the box to get rid of any doubt and to prove the outcome one way or another”. Agreed.

    Good post.

  7. If you open the box, and the cat is still warm, you know he was alive.

    but one thing is sure: the cat wil be death after you open the box.

  8. I am sorry you have completely failed with the concept of this thought experiment reducing the superposition eigenstates to a single eigenstate via the monitoring.

    If you had ever read the initial version of the experiment as presented in Die Naturwissenschaften you would know the cat is pined such that it could not interfere with the mechanism or poison.

    You also fail with the 1hour limit reducing much of your questions to be irrelevant such as lead poisoning.

    Please consider fixing this blog. It is one of the first google image results and I fear many people will get the wrong idea about this experiment after reading your post.

    • Thanks for your comment and pointing out my failures.

      I was looking at Schrödinger’s Cat thought experiment from a Software Testing perspective and not from a Science perspective. The point I was trying to get across is that you can make many assumptions around the state of an object, but without looking, or Testing it, you cannot know its true state. Again, this is from a Software Testing perspective.

      I also believe in a “Warts and all” approach to Blog writing. If the above post is wrong, it’s wrong and my reputation has suffered as a result. I will not, however, update it.

Leave a Reply

Your email address will not be published. Required fields are marked *

You may use these HTML tags and attributes: <a href="" title=""> <abbr title=""> <acronym title=""> <b> <blockquote cite=""> <cite> <code> <del datetime=""> <em> <i> <q cite=""> <strike> <strong>